Characterization of automatic treatment planning approaches in radiotherapy

Clinical implementation and evaluation of deep learning-assisted automatic radiotherapy treatment planning for lung cancer

PURPOSE

The purpose of the study is to investigate the clinical application of deep learning (DL)-assisted automatic radiotherapy planning for lung cancer.

METHODS

A DL model was developed for predicting patient-specific doses, trained and validated on a dataset of 235 patients with diverse target volumes and prescriptions. The model was integrated into clinical workflow with DL-predicted objective functions. The automatic plans were retrospectively designed for additional 50 treated manual volumetric modulated arc therapy (VMAT) plans. A comparison was made between automatic and manual plans in terms of dosimetric indexes, monitor units (MUs) and planning time. Plan quality metric (PQM) encompassing these indexes was evaluated, with higher PQM values indicating superior plan quality. Qualitative evaluations of two plans were conducted by four reviewers.

RESULTS

The PQM score was 40.7 ± 13.1 for manual plans and 40.8 ± 13.5 for automatic plans (P = 0.75). Compared to manual plans, the targets coverage and homogeneity of automatic plans demonstrated no significant difference. Manual plans exhibited better sparing for lung in V5 (difference: 1.8 ± 4.2 %, P = 0.02), whereas automatic plans showed enhanced sparing for heart in V30 (difference: 1.4 ± 4.7 %, P = 0.02) and for spinal cord in Dmax (difference: 0.7 ± 4.7 Gy, P = 0.04). The planning time and MUs of automatic plans were significantly reduced by 70.5 ± 20.0 min and 97.4 ± 82.1. Automatic plans were deemed acceptable in 88 % of the reviews (176/200).

CONCLUSIONS

The DL-assisted approach for lung cancer notably decreased planning time and MUs, while demonstrating comparable or superior quality relative to manual plans. It has the potential to provide benefit to lung cancer patients.

Clinical outcomes of pelvic bone marrow sparing radiotherapy for cervical cancer: A systematic review and meta-analysis of randomised controlled trials

Background

Concurrent chemoradiotherapy (CRT) is the standard treatment for locally advanced cervical cancer. We investigated how additional bone marrow sparing (BMS) affects the clinical outcomes.

Methods

We queried MEDLINE, Embase, Web of Science Core Collection, Google Scholar, Sinomed, CNKI, and Wanfang databases for articles published in English or Chinese between 2010/01/01 and 2023/10/31. Full-text manuscripts of prospective, randomised trials on BMS in cervical cancer patients treated with definitive or postoperative CRT were included. Risk of bias (RoB) was assessed using Cochrane Collaboration's RoB tool. Random-effects models were used for the meta-analysis.

Results

A total of 17 trials encompassing 1297 patients were included. The majority were single-centre trials (n = 1268) performed in China (n = 1128). Most trials used CT-based anatomical BMS (n = 1076). There was a comparable representation of trials in the definitive (n = 655) and postoperative (n = 582) settings, and the remaining trials included both.Twelve studies reported data on G ≥ 3 (n = 782) and G ≥ 2 (n = 754) haematologic adverse events. Both G ≥ 3 (OR 0.39; 95 % CI 0.28-0.55; p < 0.001) and G ≥ 2 (OR 0.29; 95 % CI 0.18-0.46; p < 0.001) toxicity were significantly lowered, favouring BMS. Seven studies (n = 635) reported data on chemotherapy interruptions, defined as receiving less than five cycles of cisplatin, which were significantly less frequent in patients treated with BMS (OR 0.44; 95 % CI 0.24-0.81; p = 0.016). There was no evidence of increased gastrointestinal or genitourinary toxicity.There were no signs of significant heterogeneity. Four studies were assessed as high RoB; sensitivity analyses excluding these provided comparable results for main outcomes. The main limitations include heterogeneity in BMS methodology between studies, low representation of populations most affected by cervical cancer, and insufficient data to assess survival outcomes.

Conclusions

The addition of BMS to definitive CRT in cervical cancer patients decreases hematologic toxicity and the frequency of interruptions in concurrent chemotherapy. However, data are insufficient to verify the impact on survival and disease control.

Artificial intelligence applications in prostate cancer

Artificial intelligence (AI) applications have enabled remarkable advancements in healthcare delivery. These AI tools are often aimed to improve accuracy and efficiency of histopathology assessment and diagnostic imaging interpretation, risk stratification (i.e., prognostication), and prediction of therapeutic benefit for personalized treatment recommendations. To date, multiple AI algorithms have been explored for prostate cancer to address automation of clinical workflow, integration of data from multiple domains in the decision-making process, and the generation of diagnostic, prognostic, and predictive biomarkers. While many studies remain within the pre-clinical space or lack validation, the last few years have witnessed the emergence of robust AI-based biomarkers validated on thousands of patients, and the prospective deployment of clinically-integrated workflows for automated radiation therapy design. To advance the field forward, multi-institutional and multi-disciplinary collaborations are needed in order to prospectively implement interoperable and accountable AI technology routinely in clinic.

Full daily re-optimization improves plan quality during online adaptive radiotherapy

Background and purpose

Daily online treatment plan adaptation requires a fast workflow and planning process. Current online planning consists of adaptation of a predefined reference plan, which might be suboptimal in cases of large anatomic changes. The aim of this study was to investigate plan quality differences between the current online re-planning approach and a complete re-optimization.

Material and methods

Magnetic resonance linear accelerator reference plans for ten prostate cancer patients were automatically generated using particle swarm optimization (PSO). Adapted plans were created for each fraction using (1) the current re-planning approach and (2) full PSO re-optimization and evaluated overall compliance with institutional dose-volume criteria compared to (3) clinically delivered fractions. Relative volume differences between reference and daily anatomy were assessed for planning target volumes (PTV60, PTV57.6), rectum and bladder and correlated with dose-volume results.

Results

The PSO approach showed significantly higher adherence to dose-volume criteria than the reference approach and clinical fractions (p < 0.001). In 74 % of PSO plans at most one criterion failed compared to 56 % in the reference approach and 41 % in clinical plans. A fair correlation between PTV60 D98% and relative bladder volume change was observed for the reference approach. Bladder volume reductions larger than 50 % compared to the reference plan recurrently decreased PTV60 D98% below 56 Gy.

Conclusion

Complete re-optimization maintained target coverage and organs at risk sparing even after large anatomic variations. Re-planning based on daily magnetic resonance imaging was sufficient for small variations, while large variations led to decreasing target coverage and organ-at-risk sparing.

The Future of MR-Guided Radiation Therapy

Magnetic resonance image guided radiation therapy (MRIgRT) is a relatively new technology that has already shown outcomes benefits but that has not yet reached its clinical potential. The improved soft-tissue contrast provided with MR, coupled with the immediacy of image acquisition with respect to the treatment, enables expansion of on-table adaptive protocols, currently at a cost of increased treatment complexity, use of human resources, and longer treatment slot times, which translate to decreased throughput. Many approaches are being investigated to meet these challenges, including the development of artificial intelligence (AI) algorithms to accelerate and automate much of the workflow and improved technology that parallelizes workflow tasks, as well as improvements in image acquisition speed and quality. This article summarizes limitations of current available integrated MRIgRT systems and gives an outlook about scientific developments to further expand the use of MRIgRT.

Script-based automatic radiotherapy planning for cervical cancer

BACKGROUND

This study aimed to develop fully automated script-based radiotherapy treatment plans for cervical cancer patients, and evaluate them against clinically accepted plans, as validation before clinical implementation.

MATERIAL AND METHODS

In this retrospective planning study, treatment plans for 25 locally advanced cervical cancer (LACC) patients with up to three dose levels were included. Fully automated plans were created using an in-house developed Python script in RayStation, and compared to clinically accepted manually made plans. Quantitatively, relevant dose statistics were compared, and average dose volume histograms (DVHs) were analyzed. Qualitatively, a blinded plan comparison was conducted between the clinical and automatic plans. The accuracy of treatment plan delivery was verified with the Delta4 Phantom+.

RESULTS

The quantitative evaluation showed that target coverage was acceptable for all the automatic and clinical plans. The automatic plans were significantly more conformal than the clinical plans; median of 1.03 vs. 1.12. Mean doses to almost all organs at risk (OARs) were reduced in the automatic plans, with a median reduction of between 0.6 Gy and 1.9 Gy. In the blinded plan comparison, the automatic plans were the preferred plans or of equal quality as the clinical plans in 99% of the cases. In addition, plan delivery was excellent, with a mean gamma passing rate of 99.8%. Complete script-based plans were generated in 30-45 min; about four to ten times faster than manually made plans.

CONCLUSION

The automatic plans had acceptable target coverage, lower doses to almost all OARs, more conformal dose distributions, and were predominantly preferred by the clinicians. Based on these results, our institution has implemented the script for clinical use.

Need for standardization in the use of structures in the intensity-modulated radiation therapy planning of head and neck cancers, a GORTEC study

BACKGROUND

Most radiotherapy structures contoured on CT scans during IMRT planning are defined by the ICRU, forming part of standard practice. Associated dose-volume constraints serve as parameters for dose computation algorithms to produce optimized dose maps. On the ground, however, physicists/dosimetrists routinely delineate auxiliary "non-standard" radiotherapy structures (nsRS).

MATERIALS/METHODS

From 287 patients' data, five categories of nsRS were identified. Inter-center, inter-patient variability, and temporal trends in nsRS use were investigated. Relation of nsRS with topological complexity, plan quality, calculated quality assurance (QA) and expert QA, was investigated using machine learning classification.

RESULTS

nsRS accounted for 19.2% of all structures. Average number of nsRS per patient was 8.92 ± 6.70. Variation coefficient across centers was > 70% for nsRS frequency. There was no effect of patient volume per center on averaged nsRS number between low, intermediate, and high-volume centers. No temporal trends in nsRS use were detected at the high-volume centers, except for an increase in 'forced-dose' nsRS (p = 3.08 × (10)^(-5)) at one center. Machine learning prediction accuracy including nsRS features were 0.70 ± 0.06 for topological complexity, 0.58 ± 0.05 for calculated QA and 0.72 ± 0.05 for expert QA.

CONCLUSION

Use of nsRS is frequent but heterogeneous and should be standardized further in line with ICRU initiatives in IMRT planning. Use of nsRS should be documented with respect to the need for nsRS from dose computation algorithms of treatment planning systems and IMRT machines in terms of modulation capacity and plan robustness.

A geometry and dose-volume based performance monitoring of artificial intelligence models in radiotherapy treatment planning for prostate cancer

Background and Purpose

Clinical Artificial Intelligence (AI) implementations lack ground-truth when applied on real-world data. This study investigated how combined geometrical and dose-volume metrics can be used as performance monitoring tools to detect clinically relevant candidates for model retraining.

Materials and Methods

Fifty patients were analyzed for both AI-segmentation and planning. For AI-segmentation, geometrical (Standard Surface Dice 3 mm and Local Surface Dice 3 mm) and dose-volume based parameters were calculated for two organs (bladder and anorectum) to compare AI output against the clinically corrected structure. A Local Surface Dice was introduced to detect geometrical changes in the vicinity of the target volumes, while an Absolute Dose Difference (ADD) evaluation increased focus on dose-volume related changes. AI-planning performance was evaluated using clinical goal analysis in combination with volume and target overlap metrics.

Results

The Local Surface Dice reported equal or lower values compared to the Standard Surface Dice (anorectum: (0.93 ± 0.11) vs (0.98 ± 0.04); bladder: (0.97 ± 0.06) vs (0.98 ± 0.04)). The ADD metric showed a difference of (0.9 ± 0.8)Gy for the anorectum D 1 cm 3 . The bladder D 5cm 3 reported a difference of (0.7 ± 1.5)Gy. Mandatory clinical goals were fulfilled in 90 % of the DLP plans.

Conclusions

Combining dose-volume and geometrical metrics allowed detection of clinically relevant changes, applied to both auto-segmentation and auto-planning output and the Local Surface Dice was more sensitive to local changes compared to the Standard Surface Dice. This monitoring is able to evaluate AI behavior in clinical practice and allows candidate selection for active learning.

Updating approach for lexicographic optimization-based planning to improve cervical cancer plan quality

BACKGROUND

To investigate the capability of a not-yet commercially available fully automated lexicographic optimization (LO) planning algorithm, called mCycle (Elekta AB, Stockholm, Sweden), to further improve the plan quality of an already-validated Wish List (WL) pushing on the organs-at-risk (OAR) sparing without compromising target coverage and plan delivery accuracy.

MATERIAL AND METHODS

Twenty-four mono-institutional consecutive cervical cancer Volumetric-Modulated Arc Therapy (VMAT) plans delivered between November 2019 and April 2022 (50 Gy/25 fractions) have been retrospectively selected. In mCycle the LO planning algorithm was combined with the a-priori multi-criterial optimization (MCO). Two versions of WL have been defined to reproduce manual plans (WL01), and to improve the OAR sparing without affecting minimum target coverage and plan delivery accuracy (WL02). Robust WLs have been tuned using a subset of 4 randomly selected patients. The remaining plans have been automatically re-planned by using the designed WLs. Manual plans (MP) and mCycle plans (mCP01 and mCP02) were compared in terms of dose distributions, complexity, delivery accuracy, and clinical acceptability. Two senior physicians independently performed a blind clinical evaluation, ranking the three competing plans. Furthermore, a previous defined global quality index has been used to gather into a single score the plan quality evaluation.

RESULTS

The WL tweaking requests 5 and 3 working days for the WL01 and the WL02, respectively. The re-planning took in both cases 3 working days. mCP01 best performed in terms of target coverage (PTV V95% (%): MP 98.0 [95.6-99.3], mCP01 99.2 [89.7-99.9], mCP02 96.9 [89.4-99.5]), while mCP02 showed a large OAR sparing improvement, especially in the rectum parameters (e.g., Rectum D50% (Gy): MP 41.7 [30.2-47.0], mCP01 40.3 [31.4-45.8], mCP02 32.6 [26.9-42.6]). An increase in plan complexity has been registered in mCPs without affecting plan delivery accuracy. In the blind comparisons, all automated plans were considered clinically acceptable, and mCPs were preferred over MP in 90% of cases. Globally, automated plans registered a plan quality score at least comparable to MP.

CONCLUSIONS

This study showed the flexibility of the Lexicographic approach in creating more demanding Wish Lists able to potentially minimize toxicities in RT plans.

Updating a clinical Knowledge-Based Planning prediction model for prostate radiotherapy

BACKGROUND AND PURPOSE

Clinical knowledge-based planning (KBP) models dedicated to prostate radiotherapy treatment may require periodical updates to remain relevant and to adapt to possible changes in the clinic. This study proposes a paired comparison of two different update approaches through a longitudinal analysis.

MATERIALS AND METHODS

A clinically validated KBP model for moderately hypofractionated prostate therapy was periodically updated using two approaches: one was targeted at achieving the biggest library size (M_t), while the other one at achieving the highest mean sample quality (R_t). Four subsequent updates were accomplished. The goodness, robustness and quality of the outcomes were measured and compared to those of the common ancestor. Plan quality was assessed through the Plan Quality Metric (PQM) and plan complexity was monitored.

RESULTS

Both update procedures allowed for an increase in the OARs sparing between +3.9 % and +19.2 % compared to plans generated by a human planner. Target coverage and homogeneity slightly reduced [-0.2 %;-14.7 %] while plan complexity showed only minor changes. Increasing the sample size resulted in more reliable predictions and improved goodness-of-fit, while increasing the mean sample quality improved the outcomes but slightly reduced the models reliability.

CONCLUSIONS

Repeated updates of clinical KBP models can enhance their robustness, reliability and the overall quality of automatically generated plans. The periodical expansion of the model sample accompanied by the removal of the unacceptable low quality plans should maximize the benefits of the updates while limiting the associated workload.

OpenKBP-Opt: an international and reproducible evaluation of 76 knowledge-based planning pipelines

Objective.To establish an open framework for developing plan optimization models for knowledge-based planning (KBP).Approach.Our framework includes radiotherapy treatment data (i.e. reference plans) for 100 patients with head-and-neck cancer who were treated with intensity-modulated radiotherapy. That data also includes high-quality dose predictions from 19 KBP models that were developed by different research groups using out-of-sample data during the OpenKBP Grand Challenge. The dose predictions were input to four fluence-based dose mimicking models to form 76 unique KBP pipelines that generated 7600 plans (76 pipelines × 100 patients). The predictions and KBP-generated plans were compared to the reference plans via: the dose score, which is the average mean absolute voxel-by-voxel difference in dose; the deviation in dose-volume histogram (DVH) points; and the frequency of clinical planning criteria satisfaction. We also performed a theoretical investigation to justify our dose mimicking models.Main results.The range in rank order correlation of the dose score between predictions and their KBP pipelines was 0.50-0.62, which indicates that the quality of the predictions was generally positively correlated with the quality of the plans. Additionally, compared to the input predictions, the KBP-generated plans performed significantly better (P< 0.05; one-sided Wilcoxon test) on 18 of 23 DVH points. Similarly, each optimization model generated plans that satisfied a higher percentage of criteria than the reference plans, which satisfied 3.5% more criteria than the set of all dose predictions. Lastly, our theoretical investigation demonstrated that the dose mimicking models generated plans that are also optimal for an inverse planning model.Significance.This was the largest international effort to date for evaluating the combination of KBP prediction and optimization models. We found that the best performing models significantly outperformed the reference dose and dose predictions. In the interest of reproducibility, our data and code is freely available.

Explaining the dosimetric impact of contouring errors in head and neck radiotherapy

Objective. Auto-contouring of organs at risk (OAR) is becoming more common in radiotherapy. An important issue in clinical decision making is judging the quality of the auto-contours. While recent studies considered contour quality by looking at geometric errors only, this does not capture the dosimetric impact of the errors. In this work, we studied the relationship between geometrical errors, the local dose and the dosimetric impact of the geometrical errors.Approach. For 94 head and neck patients, unmodified atlas-based auto-contours and clinically used delineations of the parotid glands and brainstem were retrieved. VMAT plans were automatically optimized on the auto-contours and evaluated on both contours. We defined the dosimetric impact on evaluation (DIE) as the difference in the dosimetric parameter of interest between the two contours. We developed three linear regression models to predict the DIE using: (1) global geometric metrics, (2) global dosimetric metrics, (3) combined local geometric and dosimetric metrics. For model (3), we next determined the minimal amount of editing information required to produce a reliable prediction. Performance was assessed by the root mean squared error (RMSE) of the predicted DIE using 5-fold cross-validation.Main results. In model (3), the median RMSE of the left parotid was 0.4 Gy using 5% of the largest editing vectors. For the right parotid and brainstem the results were 0.5 Gy using 10% and 0.4 Gy using 1% respectively. The median RMS of the DIE was 0.6 Gy, 0.7 Gy and 0.9 Gy for the left parotid, the right parotid and the brainstem, respectively. Model (3), combining local dosimetric and geometric quantities, outperformed the models that used only geometric or dosimetric information.Significance. We showed that the largest local errors plus the local dose suffice to accurately predict the dosimetric impact, opening the door to automated dosimetric QA of auto-contours.

Organ-at-risk dose prediction using a machine learning algorithm: Clinical validation and treatment planning benefit for lung SBRT

OBJECTIVE

To quantify the clinical performance of a machine learning (ML) algorithm for organ-at-risk (OAR) dose prediction for lung stereotactic body radiation therapy (SBRT) and estimate the treatment planning benefit from having upfront access to these dose predictions.

METHODS

ML models were trained using multi-center data consisting of 209 patients previously treated with lung SBRT. Two prescription levels were investigated, 50 Gy in five fractions and 54 Gy in three fractions. Models were generated using a gradient-boosted regression tree algorithm using grid searching with fivefold cross-validation. Twenty patients not included in the training set were used to test OAR dose prediction performance, ten for each prescription. We also performed blinded re-planning based on OAR dose predictions but without access to clinically delivered plans. Differences between predicted and delivered doses were assessed by root-mean square deviation (RMSD), and statistical differences between predicted, delivered, and re-planned doses were evaluated with one-way analysis of variance (ANOVA) tests.

RESULTS

ANOVA tests showed no significant differences between predicted, delivered, and replanned OAR doses (all p ≥ 0.36). The RMSD was 2.9, 3.9, 4.3, and 1.7Gy for max dose to the spinal cord, great vessels, heart, and trachea, respectively, for 50 Gy in five fractions. Average improvements of 1.0, 1.4, and 2.0 Gy were seen for spinal cord, esophagus, and trachea max doses in blinded replans compared to clinically delivered plans with 54 Gy in three fractions, and 1.8, 0.7, and 1.5 Gy, respectively, for the esophagus, heart and bronchus max doses with 50 Gy in five fractions. Target coverage was similar with an average PTV V100% of 94.7% for delivered plans compared to 97.3% for blinded re-plans for 50 Gy in five fractions, and respectively 98.4% versus 99.2% for 54 Gy in three fractions.

CONCLUSION

This study validated ML-based OAR dose prediction for lung SBRT, showing potential for improved OAR dose sparing and more consistent plan quality using dose predictions for patient-specific planning guidance.